Spinal cord injury (SCI), traumatic brain injury, stroke, multiple sclerosis, and other chronic disorders produce abnormal reflexes that impair locomotion, reach-and-grasp, and other motor functions for millions of Americans, including many Veterans. New treatments are urgently needed. Operant conditioning protocols can change spinal reflexes in rats, mice, monkeys, and people. These protocols, which are non-invasive in humans, can target beneficial plasticity to a specific reflex pathway. The reflex is elicited and the subject is rewarded if the reflex satisfies a size criterion. The subject learns to modify corticospinal control over the pathway. This control gradually changes the spinal pathway itself, and thereby triggers further beneficial plasticity elsewhere. In people with incomplete SCI, operant conditioning of the soleus H-reflex increases walking speed and reduces limping. The improvements persist; they are apparent to people in their daily lives. {Reflex conditioning in people with SCI or stroke now requires 36 one-hr sessions over 12 weeks, and is successful in only 50-70%.} Better understanding of the cortical activity that drives the reflex change should lead to better protocols that increase the reliability, magnitude, and speed of reflex conditioning, and thereby enhance its clinical value. This project seeks to identify electroencephalographic (EEG) features that reflect the crucial cortical activity, to use these features to improve the reflex conditioning protocol, {and to show that this protocol is effective in Veterans with chronic stroke.} It has two specific aims. Aim 1 will identify EEG features that correlate will the size of the H-reflex in the arm muscle flexor carpi radialis (FCR) and incorporate these features into the operant conditioning protocol. Based on human and animal data, we expect that the best feature will be sensorimotor rhythm (SMR) amplitude over contralateral sensorimotor cortex (SMC) in the 1 sec immediately before H-reflex elicitation. The new protocol will require that this EEG feature satisfy a size criterion prior to H-reflex elicitation. We expect that this new requirement will guide the person to produce, maximize, and maintain appropriate change in corticospinal influence on the reflex pathway; it will thereby increase the reliability, magnitude, and speed of H-reflex change. We will develop and validate this new protocol through studies in Veterans without neurological disease. {Aim 2 will recruit Veterans with impaired arm function due to a stroke >1 yr earlier. One group will undergo FCR H-reflex down-conditioning with the enhanced protocol; another group will undergo down- conditioning with the standard protocol. (We will down-condition the FCR H-reflex in these Veterans because it is the down-conditioning protocol that would be used clinically to reduce the hyperreflexia and/or the abnormal flexor synergy than can occur with stroke.) Because the enhanced protocol will guide the person to produce, maximize, and maintain appropriate change in corticospinal influence on the reflex pathway, we expect that its reliability will be higher, and that it will decrease the H-reflex more and more rapidly, than the standard protocol. This result will validate the enhanced protocol for people with chronic stroke.} In sum, the goal of this project is to gain new mechanistic understanding of a novel therapy and to use this knowledge to improve the therapy. {By identifying an EEG feature that reflects the cortical activity that drives the spinal plasticity underlying H-reflex change, and by showing that the feature can be used to increase the rate, magnitude, and reliability of H-reflex change in Veterans with chronic stroke, this work should augment the therapeutic value and practicality of spinal reflex conditioning.} If it is successful, it should lead to clinical trials that evaluate the ability of this new non-invasive therapy to enhance functional recovery for Veterans with stroke, spinal cord or brain injury, multiple sclerosis, or other chronic neuromuscular disorders. ! !